US20060146635A1 - Semiconductor device - Google Patents

Semiconductor device Download PDF

Info

Publication number
US20060146635A1
US20060146635A1 US11/363,060 US36306006A US2006146635A1 US 20060146635 A1 US20060146635 A1 US 20060146635A1 US 36306006 A US36306006 A US 36306006A US 2006146635 A1 US2006146635 A1 US 2006146635A1
Authority
US
United States
Prior art keywords
circuit
state
power source
clock signal
flip
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
US11/363,060
Other versions
US7254082B2 (en
Inventor
Takao Watanabe
Kunio Uchiyama
Osamu Nishii
Naohiko Irie
Hiroyuki Mizuno
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Renesas Electronics Corp
Original Assignee
Hitachi Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hitachi Ltd filed Critical Hitachi Ltd
Priority to US11/363,060 priority Critical patent/US7254082B2/en
Publication of US20060146635A1 publication Critical patent/US20060146635A1/en
Priority to US11/826,854 priority patent/US20080016383A1/en
Application granted granted Critical
Publication of US7254082B2 publication Critical patent/US7254082B2/en
Assigned to RENESAS TECHNOLOGY CORP. reassignment RENESAS TECHNOLOGY CORP. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: HITACHI, LTD.
Assigned to RENESAS ELECTRONICS CORPORATION reassignment RENESAS ELECTRONICS CORPORATION MERGER AND CHANGE OF NAME Assignors: RENESAS TECHNOLOGY CORP.
Assigned to RENESAS ELECTRONICS CORPORATION reassignment RENESAS ELECTRONICS CORPORATION CHANGE OF ADDRESS Assignors: RENESAS ELECTRONICS CORPORATION
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26Power supply means, e.g. regulation thereof
    • G06F1/32Means for saving power
    • G06F1/3203Power management, i.e. event-based initiation of a power-saving mode
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
    • G06F1/26Power supply means, e.g. regulation thereof
    • G06F1/32Means for saving power
    • G06F1/3203Power management, i.e. event-based initiation of a power-saving mode
    • G06F1/3234Power saving characterised by the action undertaken
    • G06F1/3287Power saving characterised by the action undertaken by switching off individual functional units in the computer system
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C11/00Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
    • G11C11/21Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
    • G11C11/34Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices
    • G11C11/40Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors
    • G11C11/41Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors forming static cells with positive feedback, i.e. cells not needing refreshing or charge regeneration, e.g. bistable multivibrator or Schmitt trigger
    • G11C11/413Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction
    • G11C11/417Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing, timing or power reduction for memory cells of the field-effect type
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C5/00Details of stores covered by group G11C11/00
    • G11C5/14Power supply arrangements, e.g. power down, chip selection or deselection, layout of wirings or power grids, or multiple supply levels
    • G11C5/147Voltage reference generators, voltage or current regulators; Internally lowered supply levels; Compensation for voltage drops
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • This invention relates to a semiconductor device including a semiconductor integrated circuit that is fabricated by integrating a CPU core, a logic core, a memory core, and so forth.
  • An operation speed and power consumption of a semiconductor integrated circuit have been improved owing to scaling of MOS transistors.
  • MOS transistors having a processing size of about 0.15 ⁇ m or below, however, it has become more difficult to simultaneously satisfy the speed and power performance due to scaling of a power source voltage.
  • a sub-threshold leakage current increases because a threshold voltage of the MOS transistor is set to a lower level to secure a speed at a low power source voltage.
  • CMOS circuit Most of existing semiconductor integrated circuits use a CMOS circuit ideally because an AC current with charge/discharge of a load capacitance flows through the CMOS circuit only during its operation but does not flow during standby.
  • the sub-threshold leakage current always flows as a DC current.
  • the sub-threshold leakage current becomes greater as the threshold voltage of the MOS transistor becomes lower. Therefore, in the CMOS circuit using the MOS transistor having a low threshold voltage, deterioration of power performance due to the sub-threshold leakage current reaches a level that can never be neglected.
  • the power switch cuts off the sub-threshold leakage current (hereinafter called “leakage current”) during standby of the semiconductor device.
  • the threshold voltage must be set to a lower voltage with the result that the leakage current further increases.
  • the AC current during the operation decreases in proportion to the power source voltage. In consequence, a proportion of the leakage current to the consumed current in the operation is expected to increase, and the leakage current cannot be neglected in comparison the AC current in the operation.
  • the second known technology teaches to turn off the power switch for a circuit block under a non-use state during the operation, too.
  • the following two problems arise when cut-off of the current by the power switch during the operation is accomplished.
  • the first problem is the one that the consumed current increases, on the contrary, when the power switch is frequently turned on and off. Because the entire integrated circuit does not always operate during the operation, it is possible to cut off currents to the circuit blocks under the non-use state by operating the power switch. However, when ON/OFF of the power switch is repeated within a short time, a transient current consumed for turning on the power switch may become greater than the reduction of the leakage current brought forth by cut-off of the power switch, on the contrary.
  • the second problem is the problem that the speed of the integrated circuit drops due to ON/OFF of the power switch.
  • a certain period of time hereinafter called “pre-heating time” is necessary from ON timing of the power switch to the timing when the circuit whose current is cut off can be again utilized. Therefore, the speed of the integrated circuit greatly drops in some cases.
  • the means of the invention controls the power switch by utilizing a duration time of a task of a processor (CPU) for controlling the integrated circuit. More concretely, this means looks up a task ID register and a resource management register inside the CPU and turns off a power switch for supplying currents to circuit blocks under a non-use state. Generally, a current reduction effect brought forth by cutting off the leakage current becomes effective in comparison with current consumption due to ON/OFF of the current switch when the power switch can be turned off for a time longer than several microseconds, though depending on a circuit scale.
  • the duration time of a task is generally about 1 millisecond
  • the means of the invention turns on the power switch after it is turned off and a certain time lapses away but before the next task begins.
  • this predetermined time is set to a time shorter than the balance of time acquired by subtracting the pre-heating time (time necessary for charging an artificial power source line) from the task duration time, the pre-heating time finishes before the next task beings, and the delay of the next task can be avoided.
  • FIG. 1 is a block diagram of a semiconductor device according to the invention.
  • FIG. 2 shows an example of transition of a state signal for each component
  • FIG. 3 shows input/output signals to a power control unit
  • FIG. 4 shows timing of switch control during a task
  • FIG. 5 shows timing of switch control when an interrupt processing occurs
  • FIG. 6 shows timing of signals for switch control
  • FIG. 7 shows a construction of a logic block inside a semiconductor device according to the invention.
  • FIG. 8 shows a construction of a memory block inside the semiconductor device according to the invention.
  • FIG. 1 shows a basic construction when the invention is applied to a semiconductor integrated circuit fabricated by integrating a CPU core, a DSP core, various logic cores and memory cores, and so forth, on one chip.
  • the invention is not limited to the semiconductor integrated circuit of the one-chip form shown in FIG. 1 but can also be applied to semiconductor devices of a so-called “multi-chip package” type in which a plurality of chips is integrated to one package, and to their combinations. Explanation will be hereinafter given on the one-chip form shown in FIG. 1 for ease of explanation.
  • a silicon chip CHIP includes, as main cores, a CPU core CPU, a DSP core DSP, an instruction cache I-C, a data cache D-C, an XY memory XY-M, a secondary cache 2 ND-C, a direct memory access control circuit DMAC, a peripheral IP core PER-IP and a bus control circuit BC.
  • a task ID register TID-R is a register inside the CPU core and manages each task.
  • a resource management register RM-R is a register that manages the resources (cores) used.
  • a power control unit PCU is disposed for each resource (core) as a power control unit.
  • a state decoder ST-DEC sends to the power control unit PCU a state signal ⁇ ST representing whether or not to use each resource (core).
  • I-BUS, D-BUS, P-BUS 1 and P-BUS 2 respectively represent an instruction bus, a data bus, a first peripheral bus and a second peripheral bus.
  • various cores such as the CPU core, memories and buses represent examples of the basic units that constitute a chip for digital processing.
  • these circuit blocks and basic units such as the memories will be hereinafter called “components”). Since these components are assumed to respectively have ordinary operations, explanation of their functions will be hereby omitted.
  • a power switch for each component that is not used is cut off during the operation of the semiconductor device to reduce a leakage current.
  • the state decoder ST-DEC transfers to each component a state signal ⁇ ST representing whether or not to use the component on the basis of the contents of the task ID register ID-R inside the CPU core and the resource management register RM-R managing resources (components) used.
  • the power control unit PCU in each component detects from the state signal ⁇ ST whether or not the component is used.
  • the power switch disposed inside the power control unit PCU cuts off the power source.
  • the power control unit PCU is not provided to each of the CPU core, the data cache D-C as its primary cache and the instruction cache I-C.
  • a time margin capable of cutting their power sources is too short during the operation. It is also possible not to add the power control unit PCU to other components that do not have a sufficient time margin for cutting off the power source.
  • FIG. 2 shows an example of the state signal ⁇ ST outputted from the state decoder ST-DEC.
  • sign n represents the number of components as the control object of the leakage current.
  • the state signals ⁇ ST- 1 to ⁇ ST-n represent whether or not these components are used.
  • Signs A to F in the first row represent the tasks of the CPU, and the components used are different depending on the task.
  • the state signal ⁇ ST-i is “H” (high level) in the embodiment, it means that the corresponding component is used.
  • the state signal ⁇ ST-i is “L” (low level)
  • the component 1 for example, is used for the tasks A to C and F but is not used for the tasks D and E. Therefore, it is possible for the component 1 to cut off the power switch and to reduce the leakage current in the tasks D and E.
  • FIG. 3 shows a concrete structural example of the power control unit PCU.
  • the power switch SW 1 is interposed between a power source Vdd on a high potential side and a circuit LG constituting the controlled component i.
  • the power switch SW 2 is interposed between a power source Vss on a low potential side and a circuit LG constituting the controlled component i.
  • the circuit LG is of course a circuit necessary for exploiting the function of the controlled component).
  • the drawing shows a CMOS inverter circuit as the circuit LG by way of example.
  • a power switch control circuit SWC controls the power switches SW 1 and SW 2 .
  • Symbols ⁇ ack and ⁇ RESET are signals that respectively represent the end of a pre-heating time after the power switches SW 1 and SW 2 are turned on, and a reset signal for resetting halfway the power switch control.
  • the state signal ⁇ ST-i changes to “L”
  • the power switch control circuit SWC turns off the power switches SW 1 and SW 2 . In consequence, the leakage current between the power source and the component i is cut off.
  • the power switch SW is provided to both of the power source Vdd on the high potential side and the power source Vss on the low potential side in the example shown in FIG. 3 .
  • the power switch may be provided to only one of them when no leakage path of the current exists inside the circuit constituting the component. In such a case, an occupying area becomes small and power for turning on/off the power switch can be reduced, too.
  • FIG. 4 shows the operation of the power control unit PCU of the component i when the state signal ⁇ ST-i changes to the “L” level, that is, the operation during the task duration period T-task.
  • the power control unit PCU cuts off the power switch of the component i.
  • the time T-off that means the switch-off period
  • the power control unit PCU turns on the switch (SW-on) and connects the component i to the power source.
  • the time T-off is so set as not to exceed the balance obtained by subtracting the pre-heating time T-heat of the component i from the duration time T-task of the task.
  • the pre-heating time T-heat is a time required for charging artificial power source lines Vdd′ and Vss of the circuit LG.
  • the pre-heating time varies with the construction of the component and with the circuit scale. Therefore, the T-off time is preferably set for each component.
  • a current required for driving the power switch may become greater than the effect of reducing the leakage current by turning off the power switch.
  • T-min a minimum effective period of switch-off, that is defined as the switch-off period in which consumed power reduced by the reduction of the leakage current is equal to consumed power required for driving the power switch. Since the switch must be turned on beforehand by the time corresponding to the pre-heating time as described above, too, T-off must after all satisfy the following relation: T -min ⁇ T - off ⁇ T - task ⁇ T - heat (1)
  • the value of the minimum effective time T-min varies with the leakage current and with the size of the switch but is ordinarily dozens of microseconds.
  • the pre-heating time T-heat too, is dozens of microseconds.
  • the task duration time is generally in the order of milliseconds. Therefore, when switch control is conducted on the basis of the task as in the invention, the relation (1) can be easily satisfied.
  • the minimum effective time T-min and the pre-heating time T-heat can be in advance estimated by conducting simulation through circuit analysis and by conducting calculation on the basis of circuit constants. Therefore, circuits and memories may well be so designed as to satisfy the relation (1) when deciding to which extent these members are defined as the components at the time of designing of the semiconductor device. In the example shown in FIG.
  • the component is defined for each function, but cores having mutually different functions (a plurality of peripheral IP cores, for example) may be collectively defined as one component and may be controlled by a common switch.
  • a circuit block not capable of satisfying the condition (1) is preferably kept always connected to the power source without adding the power control unit PCU.
  • the power switch is automatically turned on when the predetermined time (T-off) lapses away after the power switch is turned on.
  • This method can conceal the delay due to the pre-heating time T-heat but when this component is not used in the next task, the power switch is again turned off. Consequently, the consumed current for turning on the power switch goes to waste.
  • control may be conducted in such a fashion that the use/non-use of the component is judged beforehand with reference to the task ID register TID-R of the CPU and the power switch is not turned on when the component is not used.
  • the ⁇ RESET signal is activated to “H” with the interrupt so as to reset the sequence of power switch control and to turn on the power switch (SW-on).
  • an acknowledgment signal ⁇ ack representing the termination of pre-heating is activated to “H” at completion of pre-heating, and the next task task( 2 ) is started.
  • the interrupt occurs, the delay corresponding to the pre-heating time T-heat occurs in this way.
  • the CPU may well start the next task after the acknowledge signal ⁇ ack from all the components to be used after the interrupt are activated.
  • FIG. 6 is a time chart showing transition of each signal of a component i in continuous tasks A, B, C and D.
  • signals ⁇ SW 1 and ⁇ SW 2 for controlling the power switch have opposite polarities.
  • the power switch is ON at “L” of the signal ⁇ SW 1 and OFF at “H”.
  • the power switch is OFF at “L” of ⁇ SW 2 and ON at “H”.
  • ⁇ SW 1 is expressed as “/ ⁇ SW 1 ”.
  • the state signal ⁇ ST-i is “L”, “H”, “L” and “H” in the tasks A, B, C and D, respectively.
  • the component i changes its state to “non-use”, “use”, “non-use” and “use”, respectively, in this embodiment.
  • control shown in FIG. 4 is conducted in the task A.
  • / ⁇ SW 1 and ⁇ SW 2 are changed to “H” and “L”, respectively, and the leakage current is cut off.
  • the potentials of the signals / ⁇ SW 1 and ⁇ SW 2 are returned to the original potentials at the point earlier by the pre-heating time T-heat than the termination time of the task A and the power switch is turned on.
  • the state signal ⁇ ST-i changes to “H” and the component i is under the use state. Since pre-heating has been completed in the preceding task A, the task B can be started at this time without the occurrence of the delay time.
  • the state signal ⁇ ST-i again changes to “L” and the component i is under the non-use state. In this example, the interrupt occurs under this state and the task C changes to the task D within a shorter time than the ordinary task duration time. Control shown in FIG. 5 is conducted in this case.
  • ⁇ PRESET is inputted, the potentials of / ⁇ SW 1 and ⁇ SW 2 are changed to “L” and “H”, respectively, and the power switch is turned on.
  • the acknowledgement signal ⁇ ack changes to “H”.
  • the CPU executes the task D as the next task.
  • power control according to the invention can cut off the leakage current of the non-use components during the normal operation without inviting the delay time.
  • the interrupt occurs, too, an erroneous operation does not occur because the task shifts to a next task after termination of pre-heating is confirmed.
  • FIGS. 7 and 8 show examples of such methods.
  • FIG. 7 shows an example that uses cut-off of the leakage current described above in combination with cut-off of a clock signal in a component such as a circuit block operating in synchronism with the clock.
  • a synchronous type circuit block generally includes a flip-flop circuit FF for latching input data and a logic circuit L as its units.
  • Symbols M 1 and M 2 respectively denote a P-channel MOS transistor and an N-channel MOS transistor, and they operate as power switches.
  • the embodiment shown in FIGS. 7 and 8 uses the MOS transistors because of the advantage that these transistors can be fabricated through the same process as that of other circuits.
  • the switch may be formed of other means such as a mechanical switch disposed outside the semiconductor chip, for example.
  • a gate G 1 is a circuit for controlling whether the clock signal CLK is inputted or cut off and a gate G 2 is a circuit for preventing the output signal of the logic circuit from entering a floating state when the power switch is turned off. These circuits G 1 and G 2 may be provided to the power control unit PCU. The basic operation is explained below.
  • the gate G 1 cuts off the input of the clock signal CLK to the flip-flop FF.
  • the switch control signals / ⁇ SW 1 and ⁇ SW 2 change to the “H” level and the “L” level, respectively as described above and the power switch is cut off. Since both of the clock input and the leakage current are cut off in this way, the effect of reducing consumed power is great.
  • the drawings show only one flip-flop FF to which the clock is inputted, a large number of flip-flops exist in practice. Therefore, the power reduction effect is great by cutting off the clock input.
  • the state signal ⁇ fST-i changes to the “L” level due to the gate G 2 , the output OUT of the logic circuit is fixed.
  • FIG. 8 shows an example of the application of the invention to a memory such as a secondary cache.
  • symbols CDEC, TGEN, DEC, ARY and AMP-IO respectively represent a command decoder, a timing generator, a decoder, a memory array, and a sense amplifier-input/output circuit.
  • Symbols ADD, CMD, DI and DO respectively represent an address, a command, input data and output data. These symbols are handled as the signals themselves in some cases and as their input lines or output lines in other cases for the sake of explanation.
  • the drawing depicts a plurality of signal lines as one line to simplify illustration.
  • Symbols M 1 , M 2 , M 11 , M 12 , M 21 and M 22 represent power switches each comprising an MOS transistor.
  • M 1 and M 2 are the power switches for cutting off all the components from the power source
  • M 11 and M 21 are power switches for cutting off the decoder DEC from the power source
  • M 12 and M 22 are power switches for cutting off the sense amplifier and the input/output circuit AMP-IO from the power source.
  • the power switch for cutting off the component from the power source and the power switch for selectively cutting off the circuit elements of the component from the power source are provided. Therefore, the leakage currents of the entire components can be cut off when the components are under the non-use state and moreover, when the components are under the use state, too, the element circuit or circuits can be selectively cut off from the power source. Consequently, the reduction effect of the leakage current can be made greater. More concretely, the MOS transistors M 1 and M 2 are turned off when the state signal ⁇ ST-i is at the “L” level.
  • the MOS transistors M 1 and M 2 are ON whereas the MOS transistors M 11 and M 21 or the MOS transistor M 12 and M 22 are turned off when the decoder DEC and the sense amplifier-input/output circuit AMP-IO need not be activated.
  • the time from input and subsequent read or write of an address or command to input of a next address or command is called a “cycle time”.
  • the decoder DEC and the sense amplifier-input/output circuit AMP-IO do not always operate during this cycle time but operate for only a predetermined time that is managed by the timing generator TGEN. Therefore, the power switch can cut off the leakage current during the non-use period under control of this timing generator TGEN.
  • the memory array ARY is not discretely equipped with the switch.
  • the secondary cache is constituted with volatile memory cells such as SRAM, data will be lost if the power source is cut off. In this case, too, the current can be reduced to such an extent that the data is not lost without completely cutting off the power source. Alternatively, the data can be persevered even when the current is cut off by use of non-volatile memory cells such as ferroelectric memories or magnetic memories for the memory cells.
  • the switches M 1 and M 2 for cutting off the leakage current of the entire memories cut off the power source inclusive of that of the memory array but this operation is limited to the case where the data of the memory may well be lost with the change of the tasks.
  • the memory array ARY When the data must be preserved with the change of the tasks, a construction in which the memory array ARY is kept always connected to the power source may be employed. Further, the element circuits to be discretely equipped with the switch need not be limited to the decoder DEC and the sense amplifier-input/output circuit AMP-IO but may include other element components. Further, various other modifications may be made in the invention, whenever necessary.
  • the invention can thus reduce effectively the leakage current during the operation state of the integrated circuit.

Abstract

When a leakage current of a circuit block under a non-use state is reduced by means of a power switch, frequent ON/OFF operations of the switch within a short time invite an increase of consumed power, on the contrary. Because a pre-heating time is necessary from turn-on of the switch till the circuit block becomes usable, control of the switch during an operation deteriorates a processing time of a semiconductor device. The switch is ON/OFF-controlled with a task duration time of a CPU core for controlling logic circuits and memory cores as a unit. After the switch is turned off, the switch is again turned on before termination of the task in consideration of the pre-heating time.

Description

    BACKGROUND OF THE INVENTION
  • This invention relates to a semiconductor device including a semiconductor integrated circuit that is fabricated by integrating a CPU core, a logic core, a memory core, and so forth.
  • An operation speed and power consumption of a semiconductor integrated circuit have been improved owing to scaling of MOS transistors. In the latest semiconductor integrated circuits using MOS transistors having a processing size of about 0.15 μm or below, however, it has become more difficult to simultaneously satisfy the speed and power performance due to scaling of a power source voltage. For, a sub-threshold leakage current increases because a threshold voltage of the MOS transistor is set to a lower level to secure a speed at a low power source voltage.
  • Most of existing semiconductor integrated circuits use a CMOS circuit ideally because an AC current with charge/discharge of a load capacitance flows through the CMOS circuit only during its operation but does not flow during standby. However, the sub-threshold leakage current always flows as a DC current. The sub-threshold leakage current becomes greater as the threshold voltage of the MOS transistor becomes lower. Therefore, in the CMOS circuit using the MOS transistor having a low threshold voltage, deterioration of power performance due to the sub-threshold leakage current reaches a level that can never be neglected.
  • As a known technology for reducing this sub-threshold leakage current, a system (first known technology) has been proposed that interposes a power switch using an MOS transistor having a high threshold voltage between a circuit and a power source line and keeps this MOS transistor OFF during standby, in the Preliminary Report of International Solid State Circuits Conference [ISSCC], p. 192-193, San Francisco, US, February, 1998. A system described in JP-A-10-208473 (second known technology) corresponding to the U.S. Pat. No. 5,970,018 is another example of this type.
  • SUMMARY OF THE INVENTION
  • According to the first known technology, the power switch cuts off the sub-threshold leakage current (hereinafter called “leakage current”) during standby of the semiconductor device.
  • As miniaturization of MOS transistors further proceeds and the power source voltage becomes as low as 1 V or below, for example, the threshold voltage must be set to a lower voltage with the result that the leakage current further increases. On the other hand, the AC current during the operation decreases in proportion to the power source voltage. In consequence, a proportion of the leakage current to the consumed current in the operation is expected to increase, and the leakage current cannot be neglected in comparison the AC current in the operation.
  • The second known technology teaches to turn off the power switch for a circuit block under a non-use state during the operation, too. However, the following two problems arise when cut-off of the current by the power switch during the operation is accomplished.
  • The first problem is the one that the consumed current increases, on the contrary, when the power switch is frequently turned on and off. Because the entire integrated circuit does not always operate during the operation, it is possible to cut off currents to the circuit blocks under the non-use state by operating the power switch. However, when ON/OFF of the power switch is repeated within a short time, a transient current consumed for turning on the power switch may become greater than the reduction of the leakage current brought forth by cut-off of the power switch, on the contrary.
  • The second problem is the problem that the speed of the integrated circuit drops due to ON/OFF of the power switch. A certain period of time (hereinafter called “pre-heating time”) is necessary from ON timing of the power switch to the timing when the circuit whose current is cut off can be again utilized. Therefore, the speed of the integrated circuit greatly drops in some cases.
  • The invention employs the following means to solve the problems described above. To solve the first problem, the means of the invention controls the power switch by utilizing a duration time of a task of a processor (CPU) for controlling the integrated circuit. More concretely, this means looks up a task ID register and a resource management register inside the CPU and turns off a power switch for supplying currents to circuit blocks under a non-use state. Generally, a current reduction effect brought forth by cutting off the leakage current becomes effective in comparison with current consumption due to ON/OFF of the current switch when the power switch can be turned off for a time longer than several microseconds, though depending on a circuit scale. In contrast, since the duration time of a task is generally about 1 millisecond, the problem described above can be solved when the power switch is managed in the duration time of the task. To solve the second problem, the means of the invention turns on the power switch after it is turned off and a certain time lapses away but before the next task begins. When this predetermined time is set to a time shorter than the balance of time acquired by subtracting the pre-heating time (time necessary for charging an artificial power source line) from the task duration time, the pre-heating time finishes before the next task beings, and the delay of the next task can be avoided.
  • Other objects, features and advantages of the invention will become apparent from the following description of the embodiments of the invention taken in conjunction with the accompanying drawings.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • FIG. 1 is a block diagram of a semiconductor device according to the invention;
  • FIG. 2 shows an example of transition of a state signal for each component;
  • FIG. 3 shows input/output signals to a power control unit;
  • FIG. 4 shows timing of switch control during a task;
  • FIG. 5 shows timing of switch control when an interrupt processing occurs;
  • FIG. 6 shows timing of signals for switch control;
  • FIG. 7 shows a construction of a logic block inside a semiconductor device according to the invention; and
  • FIG. 8 shows a construction of a memory block inside the semiconductor device according to the invention.
  • DESCRIPTION OF THE EMBODIMENTS
  • Preferred embodiments of the invention will be hereinafter explained with reference to the accompanying drawings.
  • FIG. 1 shows a basic construction when the invention is applied to a semiconductor integrated circuit fabricated by integrating a CPU core, a DSP core, various logic cores and memory cores, and so forth, on one chip. However, the invention is not limited to the semiconductor integrated circuit of the one-chip form shown in FIG. 1 but can also be applied to semiconductor devices of a so-called “multi-chip package” type in which a plurality of chips is integrated to one package, and to their combinations. Explanation will be hereinafter given on the one-chip form shown in FIG. 1 for ease of explanation.
  • Referring to FIG. 1, a silicon chip CHIP includes, as main cores, a CPU core CPU, a DSP core DSP, an instruction cache I-C, a data cache D-C, an XY memory XY-M, a secondary cache 2ND-C, a direct memory access control circuit DMAC, a peripheral IP core PER-IP and a bus control circuit BC. A task ID register TID-R is a register inside the CPU core and manages each task. A resource management register RM-R is a register that manages the resources (cores) used. A power control unit PCU is disposed for each resource (core) as a power control unit. A state decoder ST-DEC sends to the power control unit PCU a state signal φST representing whether or not to use each resource (core).
  • Further, I-BUS, D-BUS, P-BUS1 and P-BUS2 respectively represent an instruction bus, a data bus, a first peripheral bus and a second peripheral bus.
  • Among the constituent units shown in FIG. 1, various cores such as the CPU core, memories and buses represent examples of the basic units that constitute a chip for digital processing. (These circuit blocks and basic units such as the memories will be hereinafter called “components”). Since these components are assumed to respectively have ordinary operations, explanation of their functions will be hereby omitted. In the invention, a power switch for each component that is not used is cut off during the operation of the semiconductor device to reduce a leakage current. Initially, the state decoder ST-DEC transfers to each component a state signal φST representing whether or not to use the component on the basis of the contents of the task ID register ID-R inside the CPU core and the resource management register RM-R managing resources (components) used. The power control unit PCU in each component detects from the state signal φST whether or not the component is used. When the component is not used, the power switch disposed inside the power control unit PCU cuts off the power source.
  • Incidentally, in the construction shown in FIG. 1, the power control unit PCU is not provided to each of the CPU core, the data cache D-C as its primary cache and the instruction cache I-C. For, a time margin capable of cutting their power sources is too short during the operation. It is also possible not to add the power control unit PCU to other components that do not have a sufficient time margin for cutting off the power source.
  • The power control operation in the invention will be explained with reference to FIG. 2 to FIG. 6. FIG. 2 shows an example of the state signal φST outputted from the state decoder ST-DEC. In this embodiment, sign n represents the number of components as the control object of the leakage current. The state signals φST-1 to φST-n represent whether or not these components are used. Signs A to F in the first row represent the tasks of the CPU, and the components used are different depending on the task. When the state signal φST-i is “H” (high level) in the embodiment, it means that the corresponding component is used. When the state signal φST-i is “L” (low level), it means that the corresponding component is not used. The component 1, for example, is used for the tasks A to C and F but is not used for the tasks D and E. Therefore, it is possible for the component 1 to cut off the power switch and to reduce the leakage current in the tasks D and E.
  • As described above, the state signal φST is transferred to the power control unit PCU in each component and the power control unit controls the power switch. FIG. 3 shows a concrete structural example of the power control unit PCU. The power switch SW1 is interposed between a power source Vdd on a high potential side and a circuit LG constituting the controlled component i. The power switch SW2 is interposed between a power source Vss on a low potential side and a circuit LG constituting the controlled component i. (Though the drawing shows only one inverter to simplify the explanation, the circuit LG is of course a circuit necessary for exploiting the function of the controlled component). The drawing shows a CMOS inverter circuit as the circuit LG by way of example. A power switch control circuit SWC controls the power switches SW1 and SW2. Symbols φack and φRESET are signals that respectively represent the end of a pre-heating time after the power switches SW1 and SW2 are turned on, and a reset signal for resetting halfway the power switch control. When the state signal φST-i changes to “L”, the power switch control circuit SWC turns off the power switches SW1 and SW2. In consequence, the leakage current between the power source and the component i is cut off.
  • Incidentally, the power switch SW is provided to both of the power source Vdd on the high potential side and the power source Vss on the low potential side in the example shown in FIG. 3. However, the power switch may be provided to only one of them when no leakage path of the current exists inside the circuit constituting the component. In such a case, an occupying area becomes small and power for turning on/off the power switch can be reduced, too.
  • The basic operation of the power control unit PCU shown in FIG. 3 will be explained with reference to FIG. 4. FIG. 4 shows the operation of the power control unit PCU of the component i when the state signal φST-i changes to the “L” level, that is, the operation during the task duration period T-task.
  • First, when the state signal φST-i reaches the “L” level, the power control unit PCU cuts off the power switch of the component i. When the time T-off (that means the switch-off period) thereafter lapses away, the power control unit PCU turns on the switch (SW-on) and connects the component i to the power source. Here, the time T-off is so set as not to exceed the balance obtained by subtracting the pre-heating time T-heat of the component i from the duration time T-task of the task. The pre-heating time T-heat is a time required for charging artificial power source lines Vdd′ and Vss of the circuit LG. Consequently, pre-heating has been finished at the point of time when this task is completed, and the occurrence of a waiting time when the component i is used for the next task can be avoided. The pre-heating time varies with the construction of the component and with the circuit scale. Therefore, the T-off time is preferably set for each component. In this case, it is possible to employ a circuit construction in which a ring oscillator circuit and a counter for adding output pulses are assembled into the power control unit PCU and are allowed to operate simultaneously with cut-off of the power switch and the power switch is turned on when the value of the counter exceeds a value set in advance for each component.
  • On the other hand, when the power switch is turned on and off within a short time as described above, a current required for driving the power switch may become greater than the effect of reducing the leakage current by turning off the power switch. Assuming that a minimum effective period of switch-off, that is defined as the switch-off period in which consumed power reduced by the reduction of the leakage current is equal to consumed power required for driving the power switch, is T-min, T-off must be greater than T-min. Since the switch must be turned on beforehand by the time corresponding to the pre-heating time as described above, too, T-off must after all satisfy the following relation:
    T-min≦T-off≦T-task−T-heat  (1)
  • The value of the minimum effective time T-min varies with the leakage current and with the size of the switch but is ordinarily dozens of microseconds. The pre-heating time T-heat, too, is dozens of microseconds. On the other hand, the task duration time is generally in the order of milliseconds. Therefore, when switch control is conducted on the basis of the task as in the invention, the relation (1) can be easily satisfied. The minimum effective time T-min and the pre-heating time T-heat can be in advance estimated by conducting simulation through circuit analysis and by conducting calculation on the basis of circuit constants. Therefore, circuits and memories may well be so designed as to satisfy the relation (1) when deciding to which extent these members are defined as the components at the time of designing of the semiconductor device. In the example shown in FIG. 1, the component is defined for each function, but cores having mutually different functions (a plurality of peripheral IP cores, for example) may be collectively defined as one component and may be controlled by a common switch. A circuit block not capable of satisfying the condition (1) is preferably kept always connected to the power source without adding the power control unit PCU.
  • In the example shown in FIG. 4, the power switch is automatically turned on when the predetermined time (T-off) lapses away after the power switch is turned on. This method can conceal the delay due to the pre-heating time T-heat but when this component is not used in the next task, the power switch is again turned off. Consequently, the consumed current for turning on the power switch goes to waste. When the waste is the problem, control may be conducted in such a fashion that the use/non-use of the component is judged beforehand with reference to the task ID register TID-R of the CPU and the power switch is not turned on when the component is not used.
  • In contrast, when an interrupt occurs, the task is given up halfway, and control to cope with this operation must be conducted. It will be assumed that an interrupt occurs when the state signal φST-i represents the non-use state for a certain component and the present task is terminated. No problem occurs if the component remains under the non-use state in the next task, too. In this case, cut-off of the power switch is kept as such and the power switch is turned on at the point corresponding to the balance obtained by subtracting the pre-heating time T-heat from the task termination time. However, when the interrupt occurs and the component is to be used in the next task, control shown in FIG. 5 becomes necessary.
  • First, the φRESET signal is activated to “H” with the interrupt so as to reset the sequence of power switch control and to turn on the power switch (SW-on). Next, an acknowledgment signal φack representing the termination of pre-heating is activated to “H” at completion of pre-heating, and the next task task(2) is started. When the interrupt occurs, the delay corresponding to the pre-heating time T-heat occurs in this way. In practice, a plurality of components exists and the pre-heating time required for each component is different. Therefore, the CPU may well start the next task after the acknowledge signal φack from all the components to be used after the interrupt are activated.
  • FIG. 6 is a time chart showing transition of each signal of a component i in continuous tasks A, B, C and D. In this time chart, signals φSW1 and φSW2 for controlling the power switch have opposite polarities. The power switch is ON at “L” of the signal φSW1 and OFF at “H”. The power switch is OFF at “L” of φSW2 and ON at “H”. To represent this power switch operation, φSW1 is expressed as “/φSW1”.
  • In FIG. 6, the state signal φST-i is “L”, “H”, “L” and “H” in the tasks A, B, C and D, respectively. In other words, the component i changes its state to “non-use”, “use”, “non-use” and “use”, respectively, in this embodiment. First, control shown in FIG. 4 is conducted in the task A. To turn off the power switch, /φSW1 and φSW2 are changed to “H” and “L”, respectively, and the leakage current is cut off. The potentials of the signals /φSW1 and φSW2 are returned to the original potentials at the point earlier by the pre-heating time T-heat than the termination time of the task A and the power switch is turned on. In the next task B, the state signal φST-i changes to “H” and the component i is under the use state. Since pre-heating has been completed in the preceding task A, the task B can be started at this time without the occurrence of the delay time. In the next task C, the state signal φST-i again changes to “L” and the component i is under the non-use state. In this example, the interrupt occurs under this state and the task C changes to the task D within a shorter time than the ordinary task duration time. Control shown in FIG. 5 is conducted in this case. Next, φPRESET is inputted, the potentials of /φSW1 and φSW2 are changed to “L” and “H”, respectively, and the power switch is turned on. When the pre-heating time T-heat thereafter lapses away, the acknowledgement signal φack changes to “H”. When all the acknowledgement signals φack of other components the state of which changes from the non-use state to the use state are switched, the CPU executes the task D as the next task. As described above, power control according to the invention can cut off the leakage current of the non-use components during the normal operation without inviting the delay time. When the interrupt occurs, too, an erroneous operation does not occur because the task shifts to a next task after termination of pre-heating is confirmed.
  • The method of cutting off the leakage current in accordance with the use state of the components has so far been described. However, greater power can be reduced in some cases depending on the components by combining this method with other methods. FIGS. 7 and 8 show examples of such methods. FIG. 7 shows an example that uses cut-off of the leakage current described above in combination with cut-off of a clock signal in a component such as a circuit block operating in synchronism with the clock. A synchronous type circuit block generally includes a flip-flop circuit FF for latching input data and a logic circuit L as its units. Symbols M1 and M2 respectively denote a P-channel MOS transistor and an N-channel MOS transistor, and they operate as power switches. Since these MOS transistors operate as the power switches, absolute values of their threshold voltages are preferably set to greater values than the absolute values of the threshold voltages of MOS transistors used in the logic circuit L. Alternatively, the amplitudes of /φSW1 and φSW2 are regulated so that a gate-to-source voltage becomes a negative potential at the time of OFF and the leakage current becomes sufficiently small when the power switch is OFF. Incidentally, the embodiment shown in FIGS. 7 and 8 uses the MOS transistors because of the advantage that these transistors can be fabricated through the same process as that of other circuits. To completely cut off the leakage current, however, the switch may be formed of other means such as a mechanical switch disposed outside the semiconductor chip, for example.
  • A gate G1 is a circuit for controlling whether the clock signal CLK is inputted or cut off and a gate G2 is a circuit for preventing the output signal of the logic circuit from entering a floating state when the power switch is turned off. These circuits G1 and G2 may be provided to the power control unit PCU. The basic operation is explained below.
  • First, when the state signal φST-i changes to the “L” level, the gate G1 cuts off the input of the clock signal CLK to the flip-flop FF. The switch control signals /φSW1 and φSW2 change to the “H” level and the “L” level, respectively as described above and the power switch is cut off. Since both of the clock input and the leakage current are cut off in this way, the effect of reducing consumed power is great. Though the drawings show only one flip-flop FF to which the clock is inputted, a large number of flip-flops exist in practice. Therefore, the power reduction effect is great by cutting off the clock input. When the state signal φfST-i changes to the “L” level due to the gate G2, the output OUT of the logic circuit is fixed. It is thus possible to prevent the leakage current (so-called “penetration current”) created when the input of the next stage enters the floating state. Next, when the time corresponding to the balance obtained by subtracting the pre-heating time from the task duration time lapses away, the power switches M1 and M2 are turned on. When the tasks are terminated, the state signal φST-i changes to the “H” level, cut-off of the clock signal CLK and fixing of the output signal are terminated, and the operation of this component starts. Here, control is conducted by inputting the state signal φST-i to the gate G1 because a long pre-heating time as required for a power switch is not necessary for inputting the clock signal. As explained above, this embodiment utilizes the task duration time, simultaneously controls the power switch and the clock input and provides the great power reduction effect.
  • FIG. 8 shows an example of the application of the invention to a memory such as a secondary cache. In FIG. 8, symbols CDEC, TGEN, DEC, ARY and AMP-IO respectively represent a command decoder, a timing generator, a decoder, a memory array, and a sense amplifier-input/output circuit. Symbols ADD, CMD, DI and DO respectively represent an address, a command, input data and output data. These symbols are handled as the signals themselves in some cases and as their input lines or output lines in other cases for the sake of explanation. The drawing depicts a plurality of signal lines as one line to simplify illustration. Symbols M1, M2, M11, M12, M21 and M22 represent power switches each comprising an MOS transistor. Among them, M1 and M2 are the power switches for cutting off all the components from the power source, M11 and M21 are power switches for cutting off the decoder DEC from the power source and M12 and M22 are power switches for cutting off the sense amplifier and the input/output circuit AMP-IO from the power source.
  • In FIG. 8, the power switch for cutting off the component from the power source and the power switch for selectively cutting off the circuit elements of the component from the power source are provided. Therefore, the leakage currents of the entire components can be cut off when the components are under the non-use state and moreover, when the components are under the use state, too, the element circuit or circuits can be selectively cut off from the power source. Consequently, the reduction effect of the leakage current can be made greater. More concretely, the MOS transistors M1 and M2 are turned off when the state signal φST-i is at the “L” level. When the state signal φST-i is at the “H” level, the MOS transistors M1 and M2 are ON whereas the MOS transistors M11 and M21 or the MOS transistor M12 and M22 are turned off when the decoder DEC and the sense amplifier-input/output circuit AMP-IO need not be activated. In memories in general, the time from input and subsequent read or write of an address or command to input of a next address or command is called a “cycle time”. The decoder DEC and the sense amplifier-input/output circuit AMP-IO do not always operate during this cycle time but operate for only a predetermined time that is managed by the timing generator TGEN. Therefore, the power switch can cut off the leakage current during the non-use period under control of this timing generator TGEN.
  • In the example shown in FIG. 8, the memory array ARY is not discretely equipped with the switch. For, since the secondary cache is constituted with volatile memory cells such as SRAM, data will be lost if the power source is cut off. In this case, too, the current can be reduced to such an extent that the data is not lost without completely cutting off the power source. Alternatively, the data can be persevered even when the current is cut off by use of non-volatile memory cells such as ferroelectric memories or magnetic memories for the memory cells. The switches M1 and M2 for cutting off the leakage current of the entire memories cut off the power source inclusive of that of the memory array but this operation is limited to the case where the data of the memory may well be lost with the change of the tasks. When the data must be preserved with the change of the tasks, a construction in which the memory array ARY is kept always connected to the power source may be employed. Further, the element circuits to be discretely equipped with the switch need not be limited to the decoder DEC and the sense amplifier-input/output circuit AMP-IO but may include other element components. Further, various other modifications may be made in the invention, whenever necessary.
  • The method of reducing the leakage current during the operation of the semiconductor device has so far been described. In this instance, it is also possible to reduce power by cutting off the power switch during standby of the semiconductor device. A delay of a certain level is permitted in most cases when the entire semiconductor device shifts from the standby state to the operation state. Therefore, the switch need not be kept ON beforehand in consideration of the pre-heating time.
  • The invention can thus reduce effectively the leakage current during the operation state of the integrated circuit.
  • It should be further understood by those skilled in the art that although the foregoing description has been made on embodiments of the invention, the invention is not limited thereto and various changes and modifications may be made without departing from the spirit of the invention and the scope of the appended claims.

Claims (7)

1-9. (canceled)
10. A semiconductor device comprising:
a power source line;
a first circuit block;
a switching element which controls electrical connection and disconnection between said power source line and at least a part of said first circuit block, wherein
said first circuit block includes
a logic circuit;
a flip-flop circuit which is controlled by a clock signal so as to latch an input signal to said logic circuit; and
a first gate circuit which controls the clock signal so as to input or not input said clock signal to said flip-flop circuit, and wherein
said semiconductor device has periods of states including:
a first state in which said logic circuit is connected to said power source line by said switching element and said clock signal is allowed to be inputted to said flip-flop by said first gate circuit;
a second state in which said logic circuit is disconnected from said power source line by said switching element and said clock signal is not allowed to be inputted to said flip-flop by said first gate circuit; and
a third state in which said logic circuit is connected to said power source line by said switching element and said clock signal is not allowed to be inputted to said flip-flop by said first gate circuit.
11. A semiconductor device according to claim 10, wherein said first circuit block has means for maintaining a voltage potential at the output of said fist circuit block, and the voltage potential at the output of said first circuit block is maintained at a certain level by said means for maintaining a voltage potential during said second state.
12. A semiconductor device according to claim 10, wherein said first circuit block transfers to the third period of state from the second state, and then transfers to the first state.
13. A semiconductor device according to claim 11, wherein said first circuit block transfers to the third state from the second state, and then transfers to the first state.
14. A semiconductor device comprising:
a power line;
first and second circuit blocks;
a first switching element which controls electrical connection and disconnection between said power source line and at least a part of said first circuit block; and
a second switching element which controls electrical connection and disconnection between said power source line and at least a part of said second circuit block, wherein each of said first and second circuit blocks includes:
a logic circuit;
a flip-flop circuit which is controlled by a clock signal so as to latch an input signal to said logic circuit and
a first gate circuit which controls the clock signal so as to input or not input said clock signal to said flip-flop circuit, wherein
said semiconductor device has a period of a state in which
said logic circuit in the second circuit block is connected to said power source line by said second switching element and said clock signal is not allowed to be inputted to said flip-flop by said first gate circuit in the second circuit block, and wherein during at least a part of said period of said state said logic circuit in the first circuit block is connected to said power source line by said first switching element and said clock signal is allowed to be inputted to said flip-flop by said first gate circuit in the first circuit block.
15. A semiconductor device comprising:
a power source line;
a plurality of circuit blocks;
switching elements which control electrical connection and disconnection between said power source line and at least a part of each of said circuit blocks, wherein
each said circuit block includes
a logic circuit;
a flip-flop circuit which is controlled by a clock signal so as to latch an input signal to said logic circuit; and
a first gate circuit which controls the clock signal so as to input or not input said clock signal to said flip-flop circuit, wherein
said semiconductor device has periods of states including:
a first state in which said logic circuit is connected to said power source line by said switching element and said clock signal is allowed to be inputted to said flip-flop by said first gate circuit;
a second state in which said logic circuit is disconnected from said power source line by said switching element and said clock signal is not allowed to be inputted to said flip-flop by said first gate circuit; and
a third state in which said logic circuit is connected to said power source line by said switching element and said clock signal is not allowed to be inputted to said flip-flop by said first gate circuit, and wherein
said plurality of blocks have different said periods of respective ones of said first, second and third states.
US11/363,060 2002-06-17 2006-02-28 Semiconductor device Expired - Fee Related US7254082B2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US11/363,060 US7254082B2 (en) 2002-06-17 2006-02-28 Semiconductor device
US11/826,854 US20080016383A1 (en) 2002-06-17 2007-07-19 Semiconductor device

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP2005-175170 2002-06-17
JP2002175170A JP2004021574A (en) 2002-06-17 2002-06-17 Semiconductor device
US10/373,042 US7023757B2 (en) 2002-06-17 2003-02-26 Semiconductor device
US11/363,060 US7254082B2 (en) 2002-06-17 2006-02-28 Semiconductor device

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US10/373,042 Continuation US7023757B2 (en) 2002-06-17 2003-02-26 Semiconductor device

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US11/826,854 Continuation US20080016383A1 (en) 2002-06-17 2007-07-19 Semiconductor device

Publications (2)

Publication Number Publication Date
US20060146635A1 true US20060146635A1 (en) 2006-07-06
US7254082B2 US7254082B2 (en) 2007-08-07

Family

ID=29728018

Family Applications (3)

Application Number Title Priority Date Filing Date
US10/373,042 Expired - Lifetime US7023757B2 (en) 2002-06-17 2003-02-26 Semiconductor device
US11/363,060 Expired - Fee Related US7254082B2 (en) 2002-06-17 2006-02-28 Semiconductor device
US11/826,854 Abandoned US20080016383A1 (en) 2002-06-17 2007-07-19 Semiconductor device

Family Applications Before (1)

Application Number Title Priority Date Filing Date
US10/373,042 Expired - Lifetime US7023757B2 (en) 2002-06-17 2003-02-26 Semiconductor device

Family Applications After (1)

Application Number Title Priority Date Filing Date
US11/826,854 Abandoned US20080016383A1 (en) 2002-06-17 2007-07-19 Semiconductor device

Country Status (2)

Country Link
US (3) US7023757B2 (en)
JP (1) JP2004021574A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190075337A (en) * 2017-12-21 2019-07-01 에스케이하이닉스 주식회사 Power gating control circuit

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003132683A (en) * 2001-10-23 2003-05-09 Hitachi Ltd Semiconductor device
JP4387815B2 (en) * 2004-01-30 2009-12-24 富士通株式会社 Serial type interface circuit, power saving method thereof, and device having serial interface
US7966511B2 (en) * 2004-07-27 2011-06-21 Intel Corporation Power management coordination in multi-core processors
KR100571646B1 (en) * 2005-03-31 2006-04-17 주식회사 하이닉스반도체 Semiconductor device with power down mode
JP4621113B2 (en) * 2005-10-28 2011-01-26 ルネサスエレクトロニクス株式会社 Semiconductor integrated circuit device
JP2007226632A (en) * 2006-02-24 2007-09-06 Denso Corp Microcomputer
JP4685040B2 (en) 2007-01-24 2011-05-18 パナソニック株式会社 Semiconductor integrated circuit and power supply control method thereof
US7868479B2 (en) * 2007-06-27 2011-01-11 Qualcomm Incorporated Power gating for multimedia processing power management
JP2009151573A (en) * 2007-12-20 2009-07-09 Fujitsu Microelectronics Ltd Semiconductor integrated circuit
US9411390B2 (en) 2008-02-11 2016-08-09 Nvidia Corporation Integrated circuit device having power domains and partitions based on use case power optimization
JP5251190B2 (en) * 2008-03-19 2013-07-31 富士通株式会社 Semiconductor device
JP5672793B2 (en) * 2010-06-24 2015-02-18 富士通株式会社 Semiconductor device and method for controlling semiconductor device
US8575993B2 (en) * 2011-08-17 2013-11-05 Broadcom Corporation Integrated circuit with pre-heating for reduced subthreshold leakage
US9471395B2 (en) 2012-08-23 2016-10-18 Nvidia Corporation Processor cluster migration techniques
US20140062561A1 (en) 2012-09-05 2014-03-06 Nvidia Corporation Schmitt receiver systems and methods for high-voltage input signals
JP6126386B2 (en) * 2013-01-22 2017-05-10 キヤノン株式会社 Information processing device
US9443564B2 (en) * 2015-01-26 2016-09-13 Semiconductor Energy Laboratory Co., Ltd. Semiconductor device, electronic component, and electronic device
JP2018180768A (en) * 2017-04-07 2018-11-15 ルネサスエレクトロニクス株式会社 Semiconductor device
KR20190063879A (en) * 2017-11-30 2019-06-10 에스케이하이닉스 주식회사 Semiconductor Apparatus
US11205469B2 (en) * 2019-07-12 2021-12-21 Micron Technology, Inc. Power domain switches for switching power reduction

Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5396635A (en) * 1990-06-01 1995-03-07 Vadem Corporation Power conservation apparatus having multiple power reduction levels dependent upon the activity of the computer system
US5532524A (en) * 1994-05-11 1996-07-02 Apple Computer, Inc. Distributed power regulation in a portable computer to optimize heat dissipation and maximize battery run-time for various power modes
US5541885A (en) * 1993-01-12 1996-07-30 Kabushiki Kaisha Toshiba High speed memory with low standby current
US5724297A (en) * 1995-12-21 1998-03-03 Hitachi, Ltd. Semiconductor integrated circuit device and method of activating the same
US5925132A (en) * 1996-01-23 1999-07-20 Canon Kabushiki Kaisha Method and apparatus for implementing power saving mode
US5959927A (en) * 1994-06-02 1999-09-28 Mitsubishi Denki Kabushiki Kaisha Semiconductor integrated circuit device having hierarchical power source arrangement
US5970018A (en) * 1996-11-20 1999-10-19 Matsushita Electrical Industrial Co., Ltd. Semiconductor integrated circuit and decode circuit for memory
US6055206A (en) * 1998-10-05 2000-04-25 Mitsubishi Denki Kabushiki Kaisha Synchronous semiconductor memory device capable of reducing power dissipation by suppressing leakage current during stand-by and in active operation
US6107836A (en) * 1992-04-14 2000-08-22 Hitachi, Ltd. Semiconductor integrated circuit device having power reduction mechanism
US6107700A (en) * 1998-05-22 2000-08-22 Mitsubishi Denki Kabushiki Kaisha Semiconductor device of hierarchical power source structure
US6212641B1 (en) * 1998-07-23 2001-04-03 Inside Out Networks Method and apparatus for improving power characteristics in a system having a reduced power mode
US6333895B1 (en) * 1999-10-07 2001-12-25 Mitsubishi Denki Kabushiki Kaisha Clock synchronous semiconductor device having a reduced clock access time
US6574150B2 (en) * 2000-07-19 2003-06-03 Oki Electric Industry Co., Ltd. Dynamic random access memory with low power consumption
US6635934B2 (en) * 2000-06-05 2003-10-21 Mitsubishi Denki Kabushiki Kaisha Semiconductor integrated circuit device operating with low power consumption
US6657911B2 (en) * 2001-10-23 2003-12-02 Hitachi, Ltd. Semiconductor device with low power consumption memory circuit
US6970019B2 (en) * 1992-04-14 2005-11-29 Masashi Horiguchi Semiconductor integrated circuit device having power reduction mechanism

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5832286A (en) * 1991-12-04 1998-11-03 Sharp Kabushiki Kaisha Power control apparatus for digital electronic device
US5615162A (en) * 1995-01-04 1997-03-25 Texas Instruments Incorporated Selective power to memory
US5574697A (en) * 1995-08-15 1996-11-12 Micron Technology, Inc. Memory device with distributed voltage regulation system
US5781490A (en) * 1996-07-03 1998-07-14 Micron Technology, Inc. Multiple staged power up of integrated circuit
JP3307866B2 (en) 1996-11-20 2002-07-24 松下電器産業株式会社 Decoding circuit
JP4071378B2 (en) * 1998-11-17 2008-04-02 株式会社ルネサステクノロジ Semiconductor circuit device
WO2000045437A1 (en) * 1999-01-26 2000-08-03 Hitachi, Ltd. Method of setting back bias of mos circuit, and mos integrated circuit
JP2005071556A (en) * 2003-08-28 2005-03-17 Renesas Technology Corp Semiconductor storage device and semiconductor integrated circuit device
US7504854B1 (en) * 2003-09-19 2009-03-17 Xilinx, Inc. Regulating unused/inactive resources in programmable logic devices for static power reduction
US7498836B1 (en) * 2003-09-19 2009-03-03 Xilinx, Inc. Programmable low power modes for embedded memory blocks
US7369815B2 (en) * 2003-09-19 2008-05-06 Qualcomm Incorporated Power collapse for a wireless terminal
JP4549711B2 (en) * 2004-03-29 2010-09-22 ルネサスエレクトロニクス株式会社 Semiconductor circuit device
KR100854908B1 (en) * 2007-03-29 2008-08-28 주식회사 하이닉스반도체 Cell array of semiconductor memory device and operating method thereof

Patent Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5396635A (en) * 1990-06-01 1995-03-07 Vadem Corporation Power conservation apparatus having multiple power reduction levels dependent upon the activity of the computer system
US6107836A (en) * 1992-04-14 2000-08-22 Hitachi, Ltd. Semiconductor integrated circuit device having power reduction mechanism
US6970019B2 (en) * 1992-04-14 2005-11-29 Masashi Horiguchi Semiconductor integrated circuit device having power reduction mechanism
US5541885A (en) * 1993-01-12 1996-07-30 Kabushiki Kaisha Toshiba High speed memory with low standby current
US5532524A (en) * 1994-05-11 1996-07-02 Apple Computer, Inc. Distributed power regulation in a portable computer to optimize heat dissipation and maximize battery run-time for various power modes
US5959927A (en) * 1994-06-02 1999-09-28 Mitsubishi Denki Kabushiki Kaisha Semiconductor integrated circuit device having hierarchical power source arrangement
US5724297A (en) * 1995-12-21 1998-03-03 Hitachi, Ltd. Semiconductor integrated circuit device and method of activating the same
US5925132A (en) * 1996-01-23 1999-07-20 Canon Kabushiki Kaisha Method and apparatus for implementing power saving mode
US5970018A (en) * 1996-11-20 1999-10-19 Matsushita Electrical Industrial Co., Ltd. Semiconductor integrated circuit and decode circuit for memory
US6107700A (en) * 1998-05-22 2000-08-22 Mitsubishi Denki Kabushiki Kaisha Semiconductor device of hierarchical power source structure
US6212641B1 (en) * 1998-07-23 2001-04-03 Inside Out Networks Method and apparatus for improving power characteristics in a system having a reduced power mode
US6055206A (en) * 1998-10-05 2000-04-25 Mitsubishi Denki Kabushiki Kaisha Synchronous semiconductor memory device capable of reducing power dissipation by suppressing leakage current during stand-by and in active operation
US6333895B1 (en) * 1999-10-07 2001-12-25 Mitsubishi Denki Kabushiki Kaisha Clock synchronous semiconductor device having a reduced clock access time
US6635934B2 (en) * 2000-06-05 2003-10-21 Mitsubishi Denki Kabushiki Kaisha Semiconductor integrated circuit device operating with low power consumption
US6574150B2 (en) * 2000-07-19 2003-06-03 Oki Electric Industry Co., Ltd. Dynamic random access memory with low power consumption
US6657911B2 (en) * 2001-10-23 2003-12-02 Hitachi, Ltd. Semiconductor device with low power consumption memory circuit

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190075337A (en) * 2017-12-21 2019-07-01 에스케이하이닉스 주식회사 Power gating control circuit
KR102420005B1 (en) * 2017-12-21 2022-07-12 에스케이하이닉스 주식회사 Power gating control circuit

Also Published As

Publication number Publication date
US20030231526A1 (en) 2003-12-18
US7254082B2 (en) 2007-08-07
JP2004021574A (en) 2004-01-22
US7023757B2 (en) 2006-04-04
US20080016383A1 (en) 2008-01-17

Similar Documents

Publication Publication Date Title
US7254082B2 (en) Semiconductor device
US6696865B2 (en) Semiconductor integrated circuit device having power reduction mechanism
JP4637512B2 (en) Semiconductor integrated circuit device
KR100309723B1 (en) Integrated circuit device including cmos tri-state drivers suitable for powerdown
US9361950B1 (en) Semiconductor device with reduced leakage current and method for manufacture of the same
JPH06237164A (en) Semiconductor integrated circuit having power reduction mechanism and electronic device using same
JPH11112297A (en) Latch circuit and semiconductor integrated circuit having the same
JP2005537768A (en) Method for reducing power consumption of state holding circuit, state holding circuit and electronic device
US20050213380A1 (en) Multiple power source-semiconductor integrated circuit
US5703825A (en) Semiconductor integrated circuit device having a leakage current reduction means
JPH0786916A (en) Semiconductor integrated circuit
JPH10322192A (en) Level-converting circuit
US6532178B2 (en) Reducing level shifter standby power consumption
JP2003045189A (en) Semiconductor memory
US7729193B2 (en) Backup for volatile state retention in the absence of primary circuit power
JP2001251180A (en) Semiconductor resistor element
US7917776B2 (en) System-on-chip including deepstop mode to reduce total leakage current and method thereof
JP2002118455A (en) Semiconductor integrated circuit
JP2003298410A (en) Semiconductor integrated circuit
US6097236A (en) Signal transfer system and method using an intermediate voltage
JP2936474B2 (en) Semiconductor integrated circuit device
KR100914553B1 (en) Semiconductor integrated circuit
US6647500B1 (en) System and method to generate a float voltage potential at output when first and second power supplies fail to supply power at the same time
JP2003264457A (en) Semiconductor integrated circuit having power reduction mechanism
JPS6218993B2 (en)

Legal Events

Date Code Title Description
STCF Information on status: patent grant

Free format text: PATENTED CASE

AS Assignment

Owner name: RENESAS TECHNOLOGY CORP., JAPAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HITACHI, LTD.;REEL/FRAME:019881/0288

Effective date: 20070329

Owner name: RENESAS TECHNOLOGY CORP.,JAPAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HITACHI, LTD.;REEL/FRAME:019881/0288

Effective date: 20070329

FEPP Fee payment procedure

Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

FPAY Fee payment

Year of fee payment: 4

AS Assignment

Owner name: RENESAS ELECTRONICS CORPORATION, JAPAN

Free format text: MERGER AND CHANGE OF NAME;ASSIGNOR:RENESAS TECHNOLOGY CORP.;REEL/FRAME:026287/0075

Effective date: 20100401

FPAY Fee payment

Year of fee payment: 8

AS Assignment

Owner name: RENESAS ELECTRONICS CORPORATION, JAPAN

Free format text: CHANGE OF ADDRESS;ASSIGNOR:RENESAS ELECTRONICS CORPORATION;REEL/FRAME:044928/0001

Effective date: 20150806

FEPP Fee payment procedure

Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

LAPS Lapse for failure to pay maintenance fees

Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

STCH Information on status: patent discontinuation

Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362

FP Lapsed due to failure to pay maintenance fee

Effective date: 20190807